Research map / Label-efficient and robust learning / Recognition and long-tail
A Generalized Earley Parser for Human Activity Parsing and Prediction
Siyuan Qi, Baoxiong Jia, Siyuan Huang, Ping Wei, Song Chun Zhu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add generalized Earley parser on classifier outputs | Bi-LSTM · CAD-120 detection | micro accuracy (P/R, %): 76.2 -> 79.4 | Improved |
| add generalized Earley parser on classifier outputs | Bi-LSTM · CAD-120 future 3s prediction | micro accuracy (P/R, %): 54.2 -> 61.5 | Improved |
| add generalized Earley parser on classifier outputs | Bi-LSTM · CAD-120 segment prediction | micro accuracy (P/R, %): 31.4 -> 72.2 | Improved |
| add generalized Earley parser on classifier outputs | Bi-LSTM · Watch-n-Patch detection | micro accuracy (P/R, %): 84.0 -> 84.8 | Improved |
| add generalized Earley parser on classifier outputs | Bi-LSTM · Watch-n-Patch future 3s prediction | micro accuracy (P/R, %): 42.1 -> 49.0 | Improved |
| add generalized Earley parser on classifier outputs | Bi-LSTM · Watch-n-Patch segment prediction | micro accuracy (P/R, %): 21.7 -> 35.6 | Improved |
| use generalized Earley parser instead of original Earley on segmented labels | ST-AOG + Earley · CAD-120 segment prediction | micro accuracy (P/R, %): 54.3 -> 72.2 | Improved |
| use generalized Earley parser instead of original Earley on segmented labels | ST-AOG + Earley · Watch-n-Patch segment prediction | micro accuracy (P/R, %): 29.4 -> 35.6 | Improved |
| use generalized Earley parser instead of original Earley on segmented labels | ST-AOG + Earley · CAD-120 detection | micro accuracy (P/R, %): 76.5 -> 79.4 | Improved |
Closest papers by meaning
- Generalized Earley Parser: Bridging Symbolic Grammars and Sequence Data for Future Prediction (ICML 2018 · similarity 0.98)
- Predicting Human Activities Using Stochastic Grammar (ICCV 2017 · similarity 0.88)
- Activity Grammars for Temporal Action Segmentation (NeurIPS 2023 · similarity 0.86)
- RHYTHM: Reasoning with Hierarchical Temporal Tokenization for Human Mobility (NeurIPS 2025 · similarity 0.82)
- Domain-Informed Label Fusion Surpasses LLMs in Free-Living Activity Classification (Student Abstract) (AAAI 2025 · similarity 0.82)
- Improving Unsupervised Constituency Parsing via Maximizing Semantic Information (ICLR 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
